A command that coordinates a two-agent development loop through shared mailbox files. One invocation runs a supervised iteration; automatic mode continues until the loop reports SHIP, BLOCKED, or NEEDS_HUMAN.

Use it to start or continue the Trio loop, run one iteration, keep iterating in the current session, or select a different loop directory.

It removes the need to manually sequence repeated agent iterations and monitor their stopping conditions. It also checks that the loop has been initialized before running.

trio — Command for coding agents by albiol2004: A command that coordinates a two-agent development loop through…

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add commands/albiol2004/trio-agent-loop/trio

Clone the repo

git clone --depth 1 https://github.com/albiol2004/trio-agent-loop

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/commands/albiol2004/trio-agent-loop/trio.svg)](https://agentmods.dev/commands/albiol2004/trio-agent-loop/trio)

Your own site

<a href="https://agentmods.dev/commands/albiol2004/trio-agent-loop/trio"><img src="https://agentmods.dev/badge/commands/albiol2004/trio-agent-loop/trio.svg" alt="Measured on agentmods" height="20"></a>

Per session 49 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 3,036 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00049	$0.03036
Opus 5	$0.00024	$0.01518
Sonnet 5	$0.00010	$0.00607
Haiku 4.5	$0.00005	$0.00304

Measured 3d ago against content hash 3de66720b1d2, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

trio scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

omp/commands/trio.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the orchestrator of the Trio two-agent loop. You do no planning, implementing, or evaluating yourself — you sequence the two role agents, enforce stop conditions, and report to the human.

Usage: /trio runs exactly one supervised iteration. /trio auto keeps iterating in the same turn until the latest loop/VERDICT.md first line is VERDICT: SHIP, VERDICT: BLOCKED, or VERDICT: NEEDS_HUMAN, honoring the stop conditions below. /trio dir=loop-<name> uses a different mailbox directory (combinable: /trio auto dir=loop-x); every loop/ reference below means that directory.

If you are coming from the Claude bundle, /trio auto replaces the old /loop /trio autonomous invocation.

0. Preflight — decide whether to run at all

Default mode: a natural-language run-the-loop request ("run the trio loop", "start it", "continue", "keep iterating") follows /trio auto semantics — full chain, no per-iteration checkpoints, until SHIP/BLOCKED/NEEDS_HUMAN. Bare /trio stays supervised (exactly one iteration).

Read loop/STATE.md. If it does not exist, tell the user to run /trio-init <goal> first and STOP.
- Collision check: if STATE.md has a mission: line, verify it still matches GOAL.md's mission sentence, and verify LOG.md's tail is consistent with the iterations you've been orchestrating. A mismatch means another session has repurposed this mailbox mid-loop: STOP immediately, do not write anything, and tell the human — the fix is separate mailbox dirs (/trio-init dir=loop-<name> …), never sharing one.
Read loop/VERDICT.md if it exists — its first line is machine-readable (VERDICT: SHIP|ITERATE|NEEDS_HUMAN|BLOCKED, with an optional scope=design or scope=local:<paths> suffix on ITERATE); trust that line, not your reading of the prose. Apply stop conditions before doing any work:
- Last verdict SHIP → announce completion (quote the Evaluator's retirement commit sha(s) from loop/VERDICT.md's commit: lines and the follow-ups), end the loop.
- Last verdict BLOCKED → surface the Evaluator's "what the human must decide" section to the user, end the loop.
- Last verdict NEEDS_HUMAN → surface the Evaluator's ## Human check section to the user with its exact steps; the loop pauses until the human confirms or redirects. Do not start another iteration in the same turn.
- iteration >= max_iterations in STATE.md → stop, summarize LOG.md, tell the user how to raise the cap in STATE.md and resume.
- Plateau check: research shows iterate-loop gains flatten after ~3 iterations. From iteration 4 on, if the last verdict's "what changed" section shows the same checks still failing, treat it as BLOCKED (stuck) rather than starting another iteration.
Otherwise increment iteration in STATE.md, set status: running, and proceed.

Read the full file on GitHub · 97 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

3d ago First seen · 97 lines · 49 tokens per session scan A 3de66720b1d2

Subscribe to this mod's changes

trio is a command published in the GitHub repository albiol2004/trio-agent-loop (2 stars, last pushed 12d ago), licensed MIT. It adds 49 tokens to every session and 3,036 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Building agents LLM agentic-coding ai-agents automation claude-code developer-tools

differential-audit

Audit EARS↔code coherence via a bidirectional differential. With no arguments, opens a scoping conversation that interprets what the user wants audited, maps it to arrows, and captures an explicit EARS selection. With one or more EARS IDs, audits those specs directly using configured defaults. Requires docs/arrows/…

jszmajda/lid · 77 tokens

arrow-maintenance

Run an audit-and-update pass on the docs/arrows/ overlay. Dispatches on project state — audits an existing overlay, bootstraps the overlay from existing LID docs, or redirects to /map-codebase or /linked-intent-dev for projects without LID.

jszmajda/lid · 59 tokens

map-codebase

Bootstrap LID in an existing (brownfield) codebase. Reads every file in scope, proposes lens-based clusterings, walks the user through reconciliation, generates skeleton LLDs/HLD/EARS and the arrow overlay, then prompts the user to flesh out the skeletons. Token-intensive by design.

jszmajda/lid · 65 tokens